Last updated:
A conversational AI system designed to help users complete tasks through natural language interaction, typically powered by a large language model.
AI assistants range from general-purpose tools to embedded enterprise assistants like Microsoft 365 Copilot. Unlike fully autonomous agents, assistants typically require human input at each step but can dramatically accelerate knowledge work. Governance considerations include data ingestion risk, parametric knowledge contamination, output verification, and the application of acceptable use policies to define what data can be shared with which assistant.
Real world example:
A legal team adopts an AI assistant for document review; the firm's AUP specifies that public-classification documents may be shared but that any document containing client PII must remain inside the firm's enterprise instance with no-training guarantees.
